Recent developments in visual-inertial and LiDAR sensors and simultaneous localization and mapping (SLAM) enable recording and digital reconstruction of the physical world. In this paper we utilize a hand-held multi-sensor platform for remotely recording and characterizing physical properties of crops on a field. The platform consists of a visual-inertial sensor, color camera and 2D LiDAR. We syncronize the data from this platform and fuse them in a standard SLAM framework to obtain a detailed model of the field environment in the form of a 3D point cloud. Such a model is then fed into semi-automated crop parameter estimation pipelines to extract the spatio-temporal variation of physical crop height and canopy cover, which may be used to support decision making for breeding and precision agriculture. We present experimental results with data collected on a winter wheat field in Eschikon, Switzerland, showing the utility of our approach towards automating variability studies in crops.